1. Field of the Invention
The present invention relates to an apparatus and method for acquiring data of fast fail memory, and more particularly, to an apparatus and method for acquiring data of fast fail memory, wherein in the operation of fail memory (FM) for failure analysis in a memory test device, failure information may be acquired in real time by comparing data recorded to a pattern generator with data read from memory under test even before completion of testing.
2. Description of the Related Art
In conventional memory test devices, memory under test (which is a device under test (DUT)) is coupled with a pattern generator for testing thereof, so that predetermined data is recorded to DUT and read again to determine whether the data is normal or not.
Korean Patent No. 540506 (Algorithm pattern producer for memory device test and memory tester using the same) discloses a technique for a data comparing part configured to compare test data with data produced from a data producing part with respect to individual clock cycles and to store information about fail memory.
As such, failure data may be analyzed by obtaining information regarding accurate failure analysis and recovery processing from not only data matching/non-matching but also information and data for failure address of the corresponding memory.
However, such information is undesirably obtained in such a manner that the data recorded in the pattern generator is compared with the data read from the DUT, and the comparison result is stored in an additional recording space, and after completion of the test, whether the data is normal or not may be determined using a failure analyzer such as a computer in the additional recording space.
As illustrated in FIG. 1, a failure analyzer 3 including a personal computer (PC) cannot conventionally access failure analysis information during storage in FM 2, in order to record failure information in a pattern generator 1.
To analyze failure, access is possible after completion of generation of the pattern in the pattern generator 1. Thus, because failure may be analyzed after completion of the pattern programming of the pattern generator 1, a sum of the patterning time of the pattern generator 1 and the failure analysis time is regarded as a final period of time taken to analyze failure.
Also, when the amount of analysis information which may be stored at once is greater than that of FM 2, it cannot be stored. Further, patterning should be performed after repetition of the same work several times.